_ Git - youtube-dl/blob - youtube_dl/extractor/einthusan.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_urlparse,
  10     compat_str,
  11 )
  12 from ..utils import (
  13     extract_attributes,
  14     ExtractorError,
  15     get_elements_by_class,
  16     urlencode_postdata,
  17 )
  18
  19
  20 class EinthusanIE(InfoExtractor):
  21     _VALID_URL = r'https?://einthusan\.tv/movie/watch/(?P<id>[0-9]+)'
  22     _TEST = {
  23         'url': 'https://einthusan.tv/movie/watch/9097/',
  24         'md5': 'ff0f7f2065031b8a2cf13a933731c035',
  25         'info_dict': {
  26             'id': '9097',
  27             'ext': 'mp4',
  28             'title': 'Ae Dil Hai Mushkil',
  29             'description': 'md5:33ef934c82a671a94652a9b4e54d931b',
  30             'thumbnail': r're:^https?://.*\.jpg$',
  31         }
  32     }
  33
  34     # reversed from jsoncrypto.prototype.decrypt() in einthusan-PGMovieWatcher.js
  35     def _decrypt(self, encrypted_data, video_id):
  36         return self._parse_json(base64.b64decode((
  37             encrypted_data[:10] + encrypted_data[-1] + encrypted_data[12:-1]
  38         ).encode('ascii')).decode('utf-8'), video_id)
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42
  43         webpage = self._download_webpage(url, video_id)
  44
  45         title = self._html_search_regex(r'<h3>([^<]+)</h3>', webpage, 'title')
  46
  47         player_params = extract_attributes(self._search_regex(
  48             r'(<section[^>]+id="UIVideoPlayer"[^>]+>)', webpage, 'player parameters'))
  49
  50         page_id = self._html_search_regex(
  51             '<html[^>]+data-pageid="([^"]+)"', webpage, 'page ID')
  52         video_data = self._download_json(
  53             'https://einthusan.tv/ajax/movie/watch/%s/' % video_id, video_id,
  54             data=urlencode_postdata({
  55                 'xEvent': 'UIVideoPlayer.PingOutcome',
  56                 'xJson': json.dumps({
  57                     'EJOutcomes': player_params['data-ejpingables'],
  58                     'NativeHLS': False
  59                 }),
  60                 'arcVersion': 3,
  61                 'appVersion': 59,
  62                 'gorilla.csrf.Token': page_id,
  63             }))['Data']
  64
  65         if isinstance(video_data, compat_str) and video_data.startswith('/ratelimited/'):
  66             raise ExtractorError(
  67                 'Download rate reached. Please try again later.', expected=True)
  68
  69         ej_links = self._decrypt(video_data['EJLinks'], video_id)
  70
  71         formats = []
  72
  73         m3u8_url = ej_links.get('HLSLink')
  74         if m3u8_url:
  75             formats.extend(self._extract_m3u8_formats(
  76                 m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native'))
  77
  78         mp4_url = ej_links.get('MP4Link')
  79         if mp4_url:
  80             formats.append({
  81                 'url': mp4_url,
  82             })
  83
  84         self._sort_formats(formats)
  85
  86         description = get_elements_by_class('synopsis', webpage)[0]
  87         thumbnail = self._html_search_regex(
  88             r'''<img[^>]+src=(["'])(?P<url>(?!\1).+?/moviecovers/(?!\1).+?)\1''',
  89             webpage, 'thumbnail url', fatal=False, group='url')
  90         if thumbnail is not None:
  91             thumbnail = compat_urlparse.urljoin(url, thumbnail)
  92
  93         return {
  94             'id': video_id,
  95             'title': title,
  96             'formats': formats,
  97             'thumbnail': thumbnail,
  98             'description': description,
  99         }